Show HN: Metapost: decentralized Reddit-like website on the blockchain
metapost.ioHello HN!
Recent events on Reddit have inspired me to resurrect a side project I've intermittently worked on for over a year: a 100% decentralized version of Reddit, running on the Algorand blockchain.
If you're familiar with platforms like Reddit or Hacker News, Metapost will feel very familiar to you. The key difference is that all data is stored on the blockchain, rendering it resistant to censorship and ensuring it will remain available to everyone forever. Metapost also supports comments, upvotes, user-created and moderated boards, and it rewards users for interacting with the platform by distributing a token. More features like mobile support, threaded comments, image posts and better moderation tools are on the roadmap.
While cryptocurrency is an excellent platform for DeFi, network like Algorand with fast and cheap transactions also makes it an excellent choice for other types of decentralized apps. Metapost is an experiment in what Web3 could enable in the future.
I would be grateful if you could give it a try: https://metapost.io/
Please keep in mind that the website is still under active development. The user interface needs some polish, a lot of functionalities are yet to be added, and you may run into occasional bugs. Also open to any feedback!
Here is the Twitter for updates: https://twitter.com/metapost_io >Why do I have to pay to post? >To store data on the blockchain, a transaction fee is required. On the Algorand network, this fee is very low, currently less than $0.01 per transaction. This minimal fee acts as a disincentive for spam. Interacting with the platform also makes you eligible to earn POST tokens. >Do I need to create an account? >No. You only need to connect your Algorand wallet. Your posts and comments will be associated with your account's address. It's hard enough to get people to post for free. Now you're going to charge $0.01/post and add the friction of a blockchain to it. This is inherently a problem with building on blockchains and why nothing useful has been built on it. It's slow. It's expensive. It's complicated for users and developers. Most blockchains have had hacks in the past. Algorand was just labeled as a security by the SEC. And most important of all, people don't value decentralized storage of information enough. I'm sure scaling an the database(s) that power Reddit is hard enough. It might be impossibly uneconomical to decentralize a Reddit clone's data on a blockchain. Thanks for the feedback! There are indeed trade-offs, and I'm aware that most people would prefer to post for free. However, I believe the 100% free model is not sustainable, and it's worth experimenting with different approaches. Companies providing these types of services tend to become hostile towards their users over time in order to pay their bills, resorting to selling personal data, holding user-generated content hostage, or cluttering their websites with ads. Charging a small transaction fee to the user might also have some advantages: it discourages spam and encourages more thoughtful posting. And we're talking about an amount that is trivial for most users. Regarding speed, Algorand provides transaction finality in under 4 seconds, which is one of the main reasons I believe it's a good fit for this project. Creating a post or comment is a bit slower than posting on Hacker News, but it still feels pretty fast. As for scalability, I'd say let's cross that bridge when we get there. Currently, Metapost is only publishing text posts to the blockchain. Hosting things like images or videos would be better suited for IPFS, with only the links being published on the chain. Ads are a perfectly legit business model. We've had free television channels for a century based on the ad model. Ad-driven model is only hated by the top percent who are willing to pay a subscription. Facebook has 2 billion users who don't care about ads. You seem to have the business model backwards. People who post there are contributing their time and knowledge. They're giving you something. There's no reason for them to pay you. That's why web 2.0 can't be replicated on blockchains and no one has ever created a successful mainstream blockchain-based web 2.0 product. In an age where platforms pay you for your content, such as Substack, I can't see how charging someone $0.01/post to write in a Reddit-like platform will have any success. Regarding speed and scale - 4 seconds is an eternity in the modern web. IPFS won't work because it's too slow and again, too expensive. > In an age where platforms pay you for your content Posting rewards the user with platform tokens. Upvoting also acts as tipping the author of the post. The economics are not 100% figured out, but I am convinced there is a way for these micro-payments to work and create a self-sustaining platform. > 4 seconds is an eternity in the modern web I disagree, but I think this is mostly a UI issue. You can have optimistic updates on the frontend while the transaction is confirming and make it feel exactly like a web2 site. I’ve worked on applications with similar latency pretty recently and the solution is often just to add more caching layers to make it feel fast and snappy for the user. >Ad-driven model is only hated by the top percent who are willing to pay a subscription. Facebook has 2 billion users who don't care about ads. You have no idea how widespread ad-bloclers are, it sounds like. Do you have plans to host image, audio, and/or video media in the future? Yes! These won't be stored directly on chain, but I plan on using IPFS with the link stored on chain similarly to what some NFT platforms are doing. At that time, does your business plan include forming an LLC and registering a designated agent at DMCA.copyright.gov? I’m planning on ultimately opensourcing the contracts and providing an easy way for anyone to host their own frontend. Why